why do hedgers trade so much?wxiong/papers/jls.pdf · why do hedgers trade so much? /s187 figure 1....

S183

[Journal of Legal Studies, vol. 43 (June 2014)]� 2014 by The University of Chicago. All rights reserved. 0047-2530/2014/4302-0021$10.00

Why Do Hedgers Trade So Much?

Ing-Haw Cheng and Wei Xiong

ABSTRACT

Futures positions of commercial hedgers in wheat, corn, soybeans, and cotton fluctuate much

more than expected output. Hedgers’ short positions are positively correlated with price

changes. Together, these observations raise doubt about the common practice of categorically

classifying trading by hedgers as hedging while classifying trading by speculators as specu-

lation, as hedgers frequently change their futures positions over time for reasons unrelated

to output fluctuations, which is arguably a form of speculation.

1. INTRODUCTION

Financial innovations such as derivatives not only facilitate risk sharingand price discovery, but critics argue that they also lead to recklessspeculation that amplifies price volatility and hinders efficient risk shar-ing (Posner and Weyl 2013). This concern has led to a debate on theregulation of financial innovation and trading of financial derivativesand warrants a benefit-cost analysis of financial regulation. In this de-bate, as well as in other broad contexts of analyzing risk sharing andtrading in financial markets, it is common to separate two groups oftraders—one group of traders with established commercial interests la-beled hedgers and another group of financial traders labeled speculators.

Perhaps because of this distinction, the debate heavily focuses onexamining the behavior and impact of speculators, with little attentionon how hedgers trade in practice. Policy prescriptions often focus on

ING-HAW CHENG is Assistant Professor of Business Administration at Dartmouth College.WEI XIONG is Professor of Economics at Princeton University. We are grateful to JoseScheinkman and Glen Weyl for helpful comments. Xiong acknowledges financial supportfrom Smith Richardson Foundation grant 2011-8691.

S184 / T H E J O U R N A L O F L E G A L S T U D I E S / V O L U M E 4 3 ( 2 ) / J U N E 2 0 1 4

the behavior of the speculator group while exempting the hedger group.1

Is this categorical treatment justified? Do hedgers trade just to hedgerisk in their commercial business? Or might there be other factors drivingtheir trading? In this paper, we systematically examine how hedgers tradein the futures markets of a set of agricultural commodities: wheat, corn,soybeans, and cotton.

Commodity futures markets offer a nice setting to examine the dis-tinction between hedgers and speculators. Futures contracts on agricul-tural commodities were early financial innovations that have a longhistory of serving farmers and commodity producers to hedge the com-modity price risk they face. The long-standing hedging pressure theoryof Keynes (1923) emphasizes the imbalance between the need of com-modity producers to short sell commodity futures contracts and the lackof interest from speculators to take the long side as a key determinantof commodity futures prices. Through the financialization of commodityfutures markets in the last decade, commodity futures became a popularasset class for portfolio investors and have attracted large inflows ofinvestment capital in the magnitude of hundreds of billions of dollarsto the long side. The large capital inflows have led to a heated debateon the role of speculation in commodity futures markets, a debate par-ticularly concerned with financial traders destabilizing commodity prices(see Cheng and Xiong [forthcoming] for a review). While this debatefocuses on financial traders, more attention on how hedgers trade is alsowarranted.

The U.S. Commodity Futures Trading Commission (CFTC) publishesdata on the aggregate position levels in its Commitments of Traders(COT) reports. By regulation, clearinghouses of commodity futures mar-kets report the end-of-day positions of traders with positions larger thancertain reporting thresholds to the CFTC, which classifies each report-able trader into several categories and reports aggregated weekly posi-tions at the group level to the public. Individual traders are distinguishedby whether they have commercial interests in each commodity (CFTC2013). For the bulk of our analysis, we focus on the behavior of pro-ducer/merchant/processor/user positions reported in the CFTC’s Dis-aggregated COT (DCOT) report and consider how these commercialhedgers trade.

1. For example, the Commodity Futures Trading Commission (CFTC) has consideredposition limits in futures markets, from which so-called bona fide hedgers may obtain ex-emptions.

W H Y D O H E D G E R S T R A D E S O M U C H ? / S185

Our analysis examines whether commercial hedgers’ trading patternsare consistent with a simple benchmark notion of hedging in which risk-averse commercial hedgers take short positions in futures to mitigatetheir endowed commodity price and output risk. We proceed in twosteps.

First, we compare the intensity of hedgers’ trading with the uncer-tainty in the aggregate output of each commodity. Intuitively, in theabsence of output uncertainty, a fixed hedging position equal to the sizeof the output would perfectly hedge the price risk faced by hedgers. Inthe presence of output uncertainty, Rolfo (1980) and Hirshleifer (1991)develop theoretical models to show that hedgers tend to underhedge asoutput is negatively correlated with price and that their hedging positionsfluctuate with expected output. Our empirical analysis shows that al-though hedgers’ futures positions are much smaller than output, thevolatility of their positions is much higher than the output volatilitymeasured by either the year-to-year output fluctuation or month-to-month fluctuation of professional output forecasts. Furthermore, al-though output uncertainty declines over the harvest season, hedgers’trading volatility remains stable throughout the year.

In the second step of our analysis, we examine what else might explainthe volatility of hedgers’ futures positions. We find that hedgers respondstrongly to changes in price. They short more futures contracts whenthe futures price rises and reduce their short position as the futures pricefalls.3 It is difficult to reconcile such trading behavior as purely that ofhedging strategies of risk-averse hedgers seeking to hedge price and out-put uncertainty. For example, if prices rise in response to a demandshock, all else equal, there is no change in the quantity of expectedoutput, yet our data suggest that hedgers’ short positions increase inresponse to the increase in price.

Taken together, the high intensity of hedgers’ trading and the sensi-tivity of their futures positions to prices are difficult to reconcile withthe view that hedgers predominantly trade to mitigate cash flow volatilityby reducing exposures. Our evidence suggests that, while the overallshort positions of hedgers in commodity futures markets do offset com-modity price risk, hedgers frequently change their positions over timefor reasons unrelated to output fluctuations. Although more elaborate

3. In a related paper, Kang, Rouwenhorst, and Tang (2013) discuss how hedgers tradefrequently and in a contrarian fashion and find that they provide liquidity to speculators.Our paper explicitly relates hedgers’ trading to output forecasts.


models of hedging may explain a portion of this behavior (Rampini,Sufi, and Viswanathan 2014), an interesting question for these modelsis whether they can also simultaneously generate the significant tradingwe observe in the data.

Overall, the distinction between hedgers and speculators based onwhether they have commercial interests or are financial traders is lessinformative than previously thought for benefit-cost analyses of financialregulation. Commercial hedgers appear to engage in both production aswell as complex trading activities traditionally viewed as the provinceof financial firms with specialized trading operations. Both types of trad-ers may be engaged in trades that contribute to price discovery or perhapsto notions of reckless speculation. The key challenge lies in distinguishingthe motive behind trades.

The paper is organized as follows. Section 2 provides some institu-tional background and describes the data used in our analysis. Section3 compares the volatility of hedgers’ position changes with the uncer-tainty in aggregate commodity output. Section 4 examines the responsesof hedgers’ futures positions to price changes. We conclude in Section5 with a discussion.

2. BACKGROUND AND DATA ON TRADERS’ POSITIONS

Centralized futures markets for agricultural commodities are some ofthe earliest markets for derivatives in the United States, dating back tothe mid-1800s and the formation of the Chicago Board of Trade. Thefutures markets for wheat, corn, soybeans, and cotton (the sample inour analysis) continue to thrive, with total open interest averaging $79billion in 2010.

Data on positions in these futures markets are collected and publishedby the CFTC. Every day, traders’ positions in excess of a specified re-porting threshold, which varies by commodity, are reported to the CFTCby exchange clearing members, futures commission merchants, and for-eign brokers. Positions are reported at the contract level (for example,December 2001 corn). These data are aggregated by the CFTC into theCOT reports and have been published weekly on Tuesdays since 2000and at a lower but regular frequency before then. Aggregate positionsin the COT account for 70–90 percent of open interest in any givenmarket.

The COT report categorizes positions into commercial and noncom-mercial on the basis of trader classifications self-reported to the CFTC.


Figure 1. Aggregate notional value of net positions for wheat, corn, soybeans, and cotton

Traders who exceed the reporting threshold are required to file CFTCForm 40, which requires them to disclose information regarding thenature of their business and whether they are using futures to hedgebusiness risk. On the basis of these forms and conversations with thetrader, the CFTC decides the appropriate classification. Since 2009, theCFTC has published a DCOT report that separates positions into thoseof producers/merchants/processors/users, swap dealers, managed money,and other reportable traders. These data are available from 2006 byusing existing 2009 classifications retroactively applied to 2006 data.The CFTC also published a Supplemental COT (SCOT) report that sep-arates positions into those of commodity index investors, commercialtraders, and noncommercial traders. Broadly, commercial (COT, SCOT)and producer/merchant/processor/user (DCOT) positions are meant toinclude the positions of traders who trade futures to hedge their businessrisk.4

Figure 1 plots the aggregated net (long minus short) notional positionvalue (computed using front-month contract prices, downloaded from

4. For a discussion of other classes of trader, see Cheng, Kirilenko, and Xiong (2013).For a detailed discussion of the explanatory notes of the Disaggregated Commitment ofTraders (DCOT) report, see CFTC (2013).


Bloomberg) for the different DCOT trader categories in the four sampleagricultural commodities.5 The figure shows that the net positions ofproducers/merchants/processors/users consistently form the short side,which suggests that producers’ net short positions are much larger thanusers’ net long positions and dominate the positions reported for thegroup as a whole. Swap dealers and managed money form the long side.Gross positions (open interest) have grown significantly since 2000, ashave the net short positions of producers and net long positions offinancial traders such as index traders and hedge funds (Cheng, Kiri-lenko, and Xiong 2013).

The U.S. Department of Agriculture (USDA) keeps close track of cropproduction in the United States and around the world. Between the 9thand 12th of every month, it publishes the World Agricultural Supplyand Demand Estimates (WASDE) report, which tracks estimated pro-duction, demand, and stocks for a large number of agricultural andlivestock products, including wheat, soybean, corn, and cotton. Thelatter three are spring-planted crops, while wheat is planted in both thewinter and spring. Beginning in May, the USDA begins forecasting cropproduction using trend yields and estimates of intended and plantedacreage.6 In June, the USDA surveys a large representative sample offarms (in 1999, over 125,000) to gather information on planted acreage,which informs subsequent production estimates. Estimates are revisedeach following month on the basis of updated surveys about farmers’expected yields through the beginning and end of fall harvest, after whichthey are surveyed about actual yields until the end of April of the nextcalendar year. Estimates from the WASDE reports thus represent boththe best real-time estimates of aggregate crop production in the UnitedStates for a coming or in-progress harvest and the best historical esti-mates of total crop production for previous harvests as well (see USDA1999).

3. OUTPUT UNCERTAINTY AND HEDGING POSITION

A hedging strategy is often referred to as buying or selling of securitiesintended to offset price fluctuations of existing positions. As a farmeris naturally exposed to price fluctuations of crops in the field, a hedging

5. Front-month contract prices are available from the Bloomberg Professional service,accessed through Bloomberg-provided terminals.

6. For wheat, estimates of winter wheat are posted in May, with spring wheat added inJuly.


strategy entails shorting commodity futures contracts to offset any pricedrop at harvest time. If there is no output uncertainty, a fixed shortposition in commodity futures with a size equal to the output wouldperfectly hedge the price uncertainty faced by the farmer. In the presenceof output uncertainty, the optimal hedging strategy is more subtle. Rolfo(1980) argues that output uncertainty leads producers to underhedgebecause output is negatively correlated with price. Indeed, by studyingprice and output uncertainty faced by cocoa producers in several coun-tries, Rolfo shows that this insight helps explain the widely observedunderhedging by farmers. Hirshleifer (1991) derives a theoretical modelto systematically examine the optimal hedging strategy with both outputand price uncertainty. It is intuitive that the optimal hedging positionfluctuates with the expected output.

We first compare the volatility of positions with the uncertainty inoutput. We measure output uncertainty in two ways, through the year-to-year fluctuations in output and through the fluctuations in themonthly output forecasts provided by the USDA in the WASDE reports.

The aggregate output of a commodity, say wheat, is determined bythe acres planted at the beginning of the season and the yield per acre.As the planting area is determined by people, the output uncertaintyfaced by farmers is mostly due to the yield. Figures 2–5 plot aggregateoutput and yield from 1960 to 2012 for wheat, corn, soybeans, andcotton. Indeed, the yield of each commodity is either the same or lessvolatile than the aggregate output, which indicates that part of the an-nual output fluctuation is due to changes in planting acreage.

Figures 6–9 plot the short positions of producers/merchants/proces-sors/users from the DCOT as well as commercial positions from theCOT in commodity futures (in output-equivalent units) in each of thesefour commodities together with the aggregate annual output.8 The fig-ures suggest that both groups’ position changes are much more volatilethan the annual output changes. While the DCOT data consistently showthat producers/merchants/processors/users are net short, the COT datashow that commercials often have near-net-zero (sometimes even long)positions, which highlights the comingling of swap dealers’ and pro-

8. To convert the output-equivalent futures position, we use the size of the contract(5,000 bushels per contract for wheat, corn, and soybeans and 50,000 pounds per contractfor cotton) as well as the metric conversions reported at the end of each World AgriculturalSupply Demand Estimates (WASDE) report (.027216 bushels per metric ton for wheat andsoybeans and .025401 bushels per metric ton for corn; cotton output is reported in millionsof 480-pound bales).

S190

Figure 2. U.S. production and yields: wheat

Figure 3. U.S. production and yields: corn

S191

Figure 4. U.S. production and yields: soybeans

Figure 5. U.S. production and yields: cotton

S192

Figure 6. Commodity output and hedgers’ futures positions: wheat

Figure 7. Commodity output and hedgers’ futures positions: corn

S193

Figure 8. Commodity output and hedgers’ futures positions: soybeans

Figure 9. Commodity output and hedgers’ futures positions: cotton


Table 1. Hedge Ratios, 2007–11

Mean SD SD/Mean

Wheat .28 .08 .29Corn .17 .04 .27Soybeans .32 .10 .32Cotton .57 .19 .34

ducers’ positions that plagues the original COT report (Cheng, Kirilenko,and Xiong 2013). For the rest of the analysis, we therefore rely on theDCOT data from 2006 onward and refer to producers/merchants/pro-cessors/users as producers given their consistently net short positions.

Table 1 shows means and standard deviations of these data in termsof hedge ratios, defined as the short position of producers in commodityfutures divided by expected output. Consistent with Rolfo (1980), hedgeratios are far less than 1, although this may be partially attributable tocomingling of users’ and producers’ positions. The average hedge ratiois roughly 28 percent in wheat, 32 percent in soybeans, 17 percent incorn, and 57 percent in cotton over this period. Notably, hedge ratiosfluctuated significantly over these years, as the standard deviations ofhedge ratios are roughly 30 percent of the mean for the four commod-ities.9

Figure 10 formalizes this notion by displaying the volatility of annualpercentage changes in producers’ futures position, output, and yield foreach of the commodities over the 5 years from 2007 to 2011, from thefirst year we can compute such changes using DCOT data through thelast year in which we have finalized ex post output.10 If producers weremaintaining fixed hedge ratios, these volatilities should be equal. How-ever, the volatility of producers’ futures position ranges from .5 to .7across the commodities, while the volatility of the actual yield changesstays in a narrow range around .07.

Next we examine patterns of changes in monthly futures positionsand expected output by month of the harvest. Although the USDA begins

9. To compute average hedge ratios across harvests, we first average hedge ratios acrossthe 52 weeks of each year and then compute averages and standard deviations of theseaverages over the harvests.

10. Annual changes in futures position were calculated by first computing the average52-week percentage change in futures position across each week of a year and then com-puting the volatility of this average across harvest years. Flipping the order of operationsand computing the 52 separate volatilities of 52-week futures changes (one for each week)and then averaging these volatilities produces even more striking results.


Figure 10. Volatility of annual percentage changes in hedgers’ futures positions, output, and yield

issuing forecasts in May based on trends, the harvest for spring cropsbegins in August for wheat and cotton and September for corn andsoybeans.11 As discussed in Section 1, each month’s report contains moreinformation about aggregate supply for the year than the previousmonth’s report. These forecasts tend to be very informative about thecoming year’s crops. Figure 11 plots the root-mean-squared error(RMSE) of the forecast for 20 years by month from harvest, scaled bythe unconditional average of the actual harvest for each commodity. Thefigure shows that the uncertainty declines monotonically as the forecastsconverge to the actual harvest. Even in the noisiest first forecast, theaverage RMSE is between 6 and 13 percent of the harvest.

Figures 12–15 plot the volatility of percentage changes in producers’futures position and the volatility of percentage changes in the monthlyforecast, again by month from harvest. Two salient observations arecommon across commodities. First, the volatility of change in the pro-ducers’ futures positions is several times larger than the volatility of thechanges in forecast. Second, the volatility of change in the producers’futures positions is large throughout the year. This volatility appears to

11. Beginning-of-harvest dates can vary by region in the United States (USDA 2010),but these months are the standardized months used by the USDA in its WASDE reportsto determine the so-called marketing year.


Figure 11. Uncertainty in output forecasts

increase during the planting season (the 2 months furthest from theharvest, just prior to the first issuance of forecasts for the next harvest,represented by the right-most two points on the graphs), as uncertaintypresumably increases with the next planting. Nonetheless, it is high be-fore then, even as output uncertainty is declining.

Figure 16 repeats these results using hedge ratios. Figure 16A showsthat the average hedge ratio across harvests is remarkably stable through-out the harvest year. In contrast, Figure 16B shows that there can belarge percentage changes in hedge ratios from month to month, as thevolatility of these changes across harvests is quite high—between 10 and50 percent.

In summary, producers’ futures positions in the four commodities areseveral times more volatile than the output uncertainty. However, thecomingling of producers’ and users’ positions can pose difficulties forinterpreting the relative volatility of changes in positions and forecasts.This leads to our next question: what cause hedgers to trade?

4. PRICE CHANGES AND HEDGING POSITION

We next focus on analyzing the correlation between changes in pro-ducers’ futures positions and prices. Figures 17–20 plots the producers’

Figure 12. Volatility of hedgers’ positions and output forecasts, 2006–11: wheat

Figure 13. Volatility of hedgers’ positions and output forecasts, 2006–11: corn

S198

Figure 14. Volatility of hedgers’ positions and output forecasts, 2006–11: soybeans

Figure 15. Volatility of hedgers’ positions and output forecasts, 2006–11: cotton

S199

Figure 16. Hedge ratios

S200

Figure 17. Hedgers’ position and commodity futures prices: wheat

Figure 18. Hedgers’ position and commodity futures prices: corn

S201

Figure 19. Hedgers’ position and commodity futures prices: soybeans

Figure 20. Hedgers’ position and commodity futures prices: cotton


short position in each of the four commodity futures together with thefutures price from January 2006 to December 2012. There is a salientpattern—producers’ short positions move in sync with the price. Thatis, as the price rises, producers increase their short position, while asthe price falls, they reduce their short position. Indeed, in contrast tothe annual volatility of changes in output, the volatility of changes inprice is on the same order of magnitude as the volatility of changes inposition, as shown in Figures 2–5.

Table 2 provides results from a regression of monthly percentagechanges in producers’ short positions on the 12-month and 1-monthpercentage changes in output forecasts and the percentage change inmonthly futures price. We include a turn-of-harvest effect to control forhow output forecasts roll over to the next harvest in May and fullyinteract this effect with the main effects of interest. We use the Neweyand West (1987) construction of the covariance matrix in computingour standard errors to allow for serial correlation. Coefficients are re-ported as standard deviations of percentage changes in positions per 1standard deviation of the right-hand-side variable.

From Table 2, we observe that, first, there is little consistent corre-lation between the monthly change of producers’ short positions andthe 12-month or 1-month change in forecasted output. Second, themonthly change in position is positively and significantly correlated withthe monthly change in futures price across all commodities. Third, thebulk of the variation in change in position is explained by changes inprice, as adding the price change term to the forecast output termsincreases the R2-value for each commodity significantly (ninefold forwheat, threefold for corn, 20-fold for soybeans, and 10-fold for cotton).

Can we explain the positive correlation between producers’ changein short positions and price changes on the basis of a pure hedgingstrategy? It is difficult to reconcile such trading behavior purely on thebasis of hedging strategies of risk-averse producers seeking to hedge priceand output uncertainty. To fix intuition, consider a representative pro-ducer who faces uncertainty in both price and output. Consider an in-crease in the price, which may arise because of a negative aggregatesupply shock or positive aggregate demand shock. In the former case,all else equal, less output needs to be hedged, yet our data suggest thathedgers increase their short positions in response to a higher price. Inthe latter case, all else equal, there is no change in the quantity ofexpected output, yet our data suggest that producers’ short positionsincrease with the price increase.

Table 2. Hedgers’ Futures Position Changes and Futures Price Changes

Wheat Corn Soybeans Cotton

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

12-Month % change inoutput forecast .014

(.13)�.022

(�.29).156

(1.61).015

(.20).009

(.08).122

(1.31)�.054

(�.53)�.243

(�2.05)*1-Month % change in

output forecast .312(.54)

.262(.78)

�.502(�5.23)**

�.011(�.06)

.205(1.22)

.384(3.19)**

�.271(�1.17)

�.056(�.26)

1-Month % change infutures price .530

(4.39)**.529

(4.35)**.628

(6.49)**.632

(4.55)**.632

(5.22)**.701

(5.30)**.461

(2.53)*.549

(3.21)**Constant .005

(.05)�.055

(�.70)�.066

(�.77)�.005

(�.04)�.049

(�.63)�.052

(�.58).184

(1.46).062

(.70).066

(.81).114

(1.19).070

(.84).028

(.33)R2 .041 .370 .379 .144 .418 .423 .023 .412 .491 .019 .185 .241

Note. Values are the results of a time-series regression at the monthly frequency of the 1-month percentage change in futures position as the dependent variableon percentage changes in output forecasts and percentage changes in futures positions. A turn-of-harvest effect for the month in which output forecasts forthe new harvest year are first issued is included and fully interacted with all other terms; these coefficients are omitted for brevity. Standard errors, in parentheses,are computed using the Newey and West (1987) construction of the covariance matrix with three lags. .t p 78

*Significant at the 5% level.**Significant at the 1% level.


Certain aspects of different hedging models may explain away a por-tion of this behavior. Negative aggregate supply shocks may put pro-ducers closer to financial distress (despite higher prices) so that they needto actively increase their hedge ratio more than that implied by thenatural passive increase following the negative quantity shock, as mightbe suggested by models of hedging such as Smith and Stulz (1985) andFroot, Scharfstein, and Stein (1993). Whether this explains the averagerelationship between price and hedging could be tested in principle byexamining whether the price reaction of trading is related to the supplyor demand component of price movements through a careful instru-mental variables analysis. Notably, however, Rampini, Sufi, and Vis-wanathan (2014) provide evidence that airline fuel hedging decreases,rather than increases, with financial distress, as hedging requires costlycollateral.

This costly collateral mechanism may induce a positive correlationbetween changes in position and prices. For example, producers mayincrease hedges in response to positive demand shocks that raise theprice and thus their net worth. An interesting question for these modelsis whether they can simultaneously generate the high degree of tradingthat we observe.

5. CONCLUSION AND DISCUSSION

Overall, it is problematic to categorically classify trading by hedgers ashedging and trading by speculators as speculation. Although hedgerstend to take short positions that hedge risk in their commercial business,on the margin, they engage in significant non-output-related trading.

One possibility is that hedgers take a view on prices just as speculatorsdo. As noted in Stulz (1996), commercial hedgers may attempt to exploitinformational advantages by trading against speculators. For example,agricultural firms may have better knowledge of local physical marketconditions across the country, as the opacity of physical markets mayinduce significant informational frictions. However, it is well known thatinformation asymmetry alone prevents, rather than leads to, trading, asin the no-trade theorem of Milgrom and Stockey (1982). Odean (1998)and Scheinkman and Xiong (2003) show that overconfidence or a beliefby each trader in an informational advantage over others helps generateexcessive trading between groups of traders. In other words, heteroge-neous beliefs induced by overconfidence in an informational advantageleads to excessive trading. Consistent with this notion of speculative


trading, another possibility is that by hedging away some of their risk,hedgers are able to speculate more heavily on the basis of their dis-agreements against speculators regarding future price movements, as inSimsek (2013).12 Finally, participants in futures markets are not pro-ducers themselves but are market makers who trade in futures marketsto hedge forward contracts written with ultimate commodity producerssuch as farmers, although our analysis implies that these producers arethemselves speculating on the price.

Any of the above possibilities raise complex questions, as marketmaking, speculation based on heterogeneous beliefs, and active tradingbased on informational advantages are at odds with the canonical notionof hedging behavior. Anecdotal evidence suggests that commercial hedg-ers speculate on prices using their position both in spot and futuresmarkets. Pleven (2012) relates stories of farmers speculating on risingcorn prices using a combination of storage and options contracts. Ag-ricultural firms such as Cargill exploit complex trading strategies thatprofit from the spread between futures and spot prices and may tilt theirexposure on the basis of information about coming shortages or over-supply in certain areas (Davis 2009). Archer-Daniels-Midland Company,a large grain processor, notes in its 2012 annual report that it “usesexchange-traded futures and exchange-traded and over-the-counter op-tions contracts as components of merchandising strategies designed toenhance margins” (Archer-Daniels-Midland Company 2012, p. 45). Al-though at odds with the canonical notion of hedging behavior, suchtrading may contribute to price discovery if it is based on genuine in-formational advantages or may lead to excessive price volatility if it isinduced by overconfidence. Further research on this issue is required.

Our analysis offers implications for benefit-cost analysis of financialregulation in two ways. First, from a conceptual point of view, ourfindings suggest the need to expand the scope of the benefit-cost analysisfrom the usual emphasis on costs brought by reckless speculation offinancial speculators to cover potential reckless speculation by marketparticipants, including hedgers with established commercial interests.Second, our findings caution against overweighting the identity of thetrader as a factor in classifying trades and instead emphasize the motiveof the trade, which may be difficult to ascertain. This caution echoes

12. The presence of heterogeneous beliefs raises challenges to welfare analysis of futuresmarket trading. See Brunnermeier, Simsek, and Xiong (2012) and Gilboa, Samuelson, andSchmeidler (2012) for recently proposed welfare criteria to analyze welfare in economicmodels with heterogeneous beliefs.


the concern raised by Cochrane (2013) and Duffie (2013) that policydistinctions based on trading motives may be more challenging than everto construct.

REFERENCES

Archer-Daniel-Midlands Company. 2012. Annual Report Pursuant to Section 13or 15(d) of the Securities Exchange Act of 1934 (Form 10-K). http://edgar.sec.gov/Archives/edgar/data/7084/000000708412000034/adm10kfy12.htm.

Brunnermeier, Markus, Alp Simsek, and Wei Xiong. 2012. A Welfare Criterionfor Models with Distorted Beliefs. Working paper. Princeton University, De-partment of Economics, Princeton, N.J.

CFTC (Commodities Futures Trading Commission). 2013. Disaggregated Explan-atory Notes. http://www.cftc.gov/MarketReports/CommitmentsofTraders/DisaggregatedExplanatoryNotes/index.htm.

Cheng, Ing-Haw, Andrei Kirilenko, and Wei Xiong. 2013. Convective Risk Flowsin Commodity Futures Markets. Working paper. Dartmouth College, TuckSchool of Business, Hanover, N.H.

Cheng, Ing-Haw, and Wei Xiong. Forthcoming. The Financialization of Com-modity Markets. Annual Review of Financial Economics.

Cochrane, John. 2013. Cost-Benefit Analysis as a Framework for Financial Reg-ulation. Working paper. University of Chicago, Booth School of Business,Chicago.

Davis, Ann. 2009. Cargill’s Inside View Helps It Buck Downturn. Wall StreetJournal, January 14. http://online.wsj.com/news/articles/SB123189501407679581.

Duffie, Darrell. 2013. Challenges to a Policy Treatment of Speculative TradingMotivated by Differences in Beliefs. Working paper. Stanford University, Grad-uate School of Business, Stanford, Calif.

Froot, Kenneth, David Scharfstein, and Jeremy Stein. 1993. Risk Management:Coordinating Corporate Investment and Financing Policies. Journal of Finance48:1629–58.

Gilboa, Itzhak, Larry Samuelson, and David Schmeidler. 2012. No-Betting ParetoDominance. Working paper. Yale University, Department of Economics, NewHaven, Conn.

Hirshleifer, David. 1991. Seasonal Patterns of Futures Hedging and the Resolutionof Output Uncertainty. Journal of Economic Theory 53:304–27.

Kang, Wenjin, K. G. Rouwenhorst, and Ke Tang. 2013. Who Provides Liquidityto Commodity Futures Markets? Working paper. National University of Sin-gapore, Department of Finance, Singapore.

Keynes, John Maynard. 1923. Some Aspects of Commodity Markets. ManchesterGuardian Commercial, European Reconstruction Series, sec. 13, pp. 784–86.

http://edgar.sec.gov/Archives/edgar/data/7084/000000708412000034/adm10kfy12.htm

http://edgar.sec.gov/Archives/edgar/data/7084/000000708412000034/adm10kfy12.htm

http://www.cftc.gov/MarketReports/CommitmentsofTraders/DisaggregatedExplanatoryNotes/index.htm

http://www.cftc.gov/MarketReports/CommitmentsofTraders/DisaggregatedExplanatoryNotes/index.htm

http://online.wsj.com/news/articles/SB123189501407679581



Milgrom, Paul, and Nancy Stokey. 1982. Information, Trade, and CommonKnowledge. Journal of Economic Theory 26:17–27.

Newey, Whitney K., and Kenneth D. West. 1987. A Simple, Positive Semi-Definite,Heteroskedasticity and Autocorrelation Consistent Covariance Matrix. Econ-ometrica 55:703–8.

Odean, Terrance. 1998. Volume, Volatility, Price, and Profit When All Traders Areabove Average. Journal of Finance 53:1887–1934.

Pleven, Liam. 2012. Behind Corn’s Squeeze Play: Farmers. Wall Street Journal,February 28. http://online.wsj.com/news/articles/SB10001424052970204778604577243491711608130.

Posner, Eric, and Eric Weyl. 2013. Benefit-Cost Analysis for Financial Regulation.American Economic Review 103:393–97.

Rampini, Adriano A., Amir Sufi, and S. Viswanathan. 2014. Dynamic Risk Man-agement. Journal of Financial Economics 111:271–96.

Rolfo, Jacques. 1980. Optimal Hedging under Price and Quantity Uncertainty:The Case of a Cocoa Producer. Journal of Political Economy 88:100–116.

Scheinkman, Jose, and Wei Xiong. 2003. Overconfidence and Speculative Bubbles.Journal of Political Economy 111:1183–1219.

Simsek, Alp. 2013. Speculation and Risk Sharing with New Financial Assets. Quar-terly Journal of Economics 128:1365–96.

Smith, Clifford, and Rene Stulz. 1985. The Determinants of Firms’ Hedging Pol-icies. Journal of Financial and Quantitative Analysis 20:391–405.

Stulz, Rene. 1996. Rethinking Risk Management. Journal of Applied CorporateFinance 9:8–24.

USDA (U.S. Department of Agriculture). 1999. Understanding USDA Crop Fore-casts. Publication No. 1554. Washington, D.C.: USDA.

———. 2010. Field Crops: Usual Planting and Harvesting Dates. AgriculturalHandbook No. 628. Washington, D.C.: USDA.



why do hedgers trade so much?wxiong/papers/jls.pdf · why do hedgers trade so much? /s187 figure 1....

Documents